Photochemical cyclohexane carbonylation cocatalyzed by d8

Soc. , 1992, 114 (1), pp 350–351. DOI: 10.1021/ja00027a049. Publication Date: January 1992. ACS Legacy Archive. Cite this:J. Am. Chem. Soc. 114, 1, ...
0 downloads 0 Views 316KB Size
J. Am. Chem. Soc 1992, 114, 350-351

350

associated with the conjugate acid of the amide base used in their formation.I2 Numerous deuterium-incorporation experiments strongly support the conjecture that such postenolization complexes remain intact in solution. Likewise, we now suggest that the conflicting reports of the a.bility of LiHMDS to enolize bulky ester and amide substrates are due to the formation of relatively stable preenolization complexes as characterized herein.

reaction rate 200-fold (q5 = 5.8 X dad, = 0.20).5 Presumably benzaldehyde produced in the benzene/cyclohexane mixtures is responsible for the increased rate of cyclohexane carbonylation.

Acknowledgment. This work was supported by the National Institutes of Health through Grants GM-35982 and a Research Career Development Award (CA-01330) to P.G.W. The X-ray equipment was purchased with assistance from an instrument grant from the NSF (CHE-8206423).

Benzophenone, acetophenone, or p-(trifluoromethy1)acetophenone have a similar effect on reaction 1. 2-Nonanone and I-nonanal do not absorb the incident irradiation (A > 340 nm) and do not affect the rate. Triphenylene and anthracene absorb 366-nm light and are good triplet sensitizers’ like the aromatic carbonyls, but do not accelerate reaction 1. Unlike the aromatic carbonyls, however, their lowest excited states are m*, not n-*, and therefore they do not abstract hydrogen from alkanes, which is a well-known reaction of photoexcited ketones and aldehydes.* The role of the organic carbonyl as the photoactive species in this system is demonstrated by experiments in which the concentration of 1 was varied. An inverse dependence of the reaction rate (drel= 4.5 X 10-2-5.4 X on [ l ] is observed over the range 0.70-7.0 mM 1 as the fraction of light absorbed by @CF3C6H,)C(0)CH3 (in competition with 1) varies in this range from 0.31 to 0.043 (&dj = 0.12 f 0.02). A similar dependence is observed by varying [Ir(PMe3)2(C0)2Cl](2; vide infra). When [2] is held constant, the rate shows a positive dependence on the concentration of acetophenone over the range 0.00604.10 M (& = 1.2 X 10-2-8.4 X as the fraction of light absorbed by the ketone increases from 0.020 to 0.25. The quantum yield is invariant (f5%) over a 10-fold intensity range (2 mM 2, 0.10 M acetophenone), arguing against the possibility of a two-photon mechanism. 1-Catalyzed benzene photocarbonylation is not accelerated by aromatic ketones. This selectivity is consistent with a proposed role of the photoexcited ketone as a hydrogen-abstraction agent since the aryl-hydrogen bond is too strong for the hydrogen to undergo abstraction. The presence of a good hydrogen donor, PhCHMeOH (1 .O M), reduces the rate of reaction 1 by 40% (2 mM 2,O.lO M acetophenone). Irradiation of a 1:l C6HI2/C6Dl2 solution of 2 (4.0 mM) and Ph2C0 (0.033 M) yields the following ratio of isotopomers: C 6 H I1CHO:C6HllCDO:C6Dl ,CHO: C,$llCDO = 70.9:12.2:14.8:2.1. The isotope effect, kH/kD= 4.9, is consistent with known isotope effects for H abstraction by photoexcited ketones9 and contrasts with kH/kD = 1.1 for 1photocatalyzed benzene carbonylation. Presuming the photoexcited ketone to cleave the alkane C-H bond, we considered that metal carbonyls other than 1 might also be effective cocatalysts. Indeed, the following metal carbonyls with d8 electronic configurations are all effective (0.10 M PhC(0)Me): RhL,(CO)CI (L = PPh,, P’Pr3),Ru(CO),(dmpe), and 2.1° In the absence of ketone, none of these complexes significantly catalyze cyclohexane carbonylation.’1,’2

Supplementary Material Available: Atomic numbering schemes and tables of crystallographic data, atomic positional parameters and thermal parameters, bond lengths and angles, and selected torsion angles for the LiHMDSltert-butyl isobutyrate complex l a and the LiHMDSltert-butyl pivalate complex l b and IR spectra of la, lb, and the LiHMDS/methyl pivalate ester complex (28 pages). Ordering information is given on any current masthead page. (12) (a) Laube, T.; Dunitz, J. D.; Seebach, D. Hela. Chim. Acta 1985, 68, 1373. (b) Wanat, R. A,; Collum, D. B.; Van Duyne, G.; Clardy, J.; DePue, R. T. J . Am. Chem. Soc. 1986, 108, 3415. (c) Buchholz, S.; Harms, K.; Massa, W.; Boche, G. Angew. Chem., Int. Ed. Engl. 1989, 28(1), 72.

Photochemical Cyclohexane Carbonylation Cocatalyzed by d8 Transition Metal Carbonyls and Aromatic Ketones and Aldehydes William T. Boese and Alan S. Goldman* Department of Chemistry, Rutgers The State University of New Jersey New Brunswick. New Jersey 08903 Received June 14, 1991 Revised Manuscript Received October 30, 1991 The catalytic functionalization of alkanes is of great current interest;’ a potentially important example is carbonylation. Rh(PR3),(CO)C1 (1, R = Me) has been found to photochemically catalyze hydrocarbon ~ a r b o n y l a t i o n . Recently ~~~ it was reported that the rate of cyclohexane carbonylation upon irradiation of a 1:1 benzene/cyclohexane solution is 10 times greater than that of a solution of 1 in pure cy~lohexane.~This observation was attributed to the inhibition “of secondary photoreactions of cyclohexanecarboxaldehyde including decarbonylation due to the filter effect of b e n ~ e n e ” . ~ We find that the initial rate of cyclohexane carbonylation (A > 340 nm) in a benzene/cyclohexane solution of 1 (7.0 mM) is less than that of a pure cyclohexane solution of 1 = 2.9 X However, added benzaldehyde (0.10 M) increases the ( I ) For reviews in the area of C-H bond activation by homogeneous transition-metal systems, see: (a) Bergman, R. G. Science (Washington, D.C.) 1984, 223, 902. (b) Crabtree, R. H. Chem. Rea. 1985,85, 245. (c) Halpern, J. Inorg. Chim. Acfa 1985,100,41. (d) Jones, W. D.; Feher, F. J. Ace. Chem. Res. 1989, 22, 91. (e) Acfiuafionand Funcfionalization of Alkanes; Hill, C., Ed.; John Wiley and Sons: New York, 1989 and references therein. (2) (a) Fisher, B. J.; Eisenberg, R. Organometallics 1983, 2, 764-767. (b) Kunin, A. J.; Eisenberg, R. J . Am. Chem. Soc. 1986, 108, 535-536. (5) Kunin, A. J.; Eisenberg, R. Organometallics 1988, 7,2124-2129. (d) Gordon, E. M.; Eisenberg, R. J . Mol. Cafal. 1988, 45, 57-71. (3) (a) Sakakura, T.; Tanaka, M. J . Chem. Soc., Chem. Commun. 1987, 758-759. (b) Sakakura, T.; Tanaka, M. Chem. Let!. 1987, 249-254. (c) Sakakura, T.; Tanaka, M. Chem. Leu. 1987, 1 1 13-1 116. (d) Sakakura, T.; Sasaki, K.; Tokunaga. Y . ;Wada, K.; Tanaka, M. Chem. Left. 1988, 155-158. (4) Sakakura, T.; Sodeyama, T.; Sasaki, K.; Wada, K.; Tanaka, M. J . Am. Chem. Soc. 1990, 112, 7221-7229.

0002-7863/92/ 15 14-350$03.00/0

CyH

+ CO

metal carbonyl hu ( A

> 340 nm)

aromalic ketone or aldehyde

CyCHO

(5) All photochemical experiments (cyclohexane solvent) were performed as described previously.6 Relative quantum yields, were obtained based on rates relative to a carbonylation reaction for which the absolute q5 was measured (0.22 mM h-’ CyCHO, 7 mM 1.0.10 M PhC(O)Me, X = 366 nm, I = 1.4 X IO-’ einstein/s, 2.0 mL, q5 = 1.0 X 10.’). Use of a Corning CS-0-52 cutoff filter (A > 345 nm) instead of a “monochromatic” (366 nm) filter combination (Corning CS-0-52 and Corning CS-7-60) yielded a rate of 0.62 mM/h. Thus q5rc, = ((d[CyCHO]/df)/(0.62 mM h-’)J X 1.0 X IO-‘ (see supplementary material for experimental details). q5Jd, is defined as moles of X product per einstein absorbed^ by the ketone or aldeLyde, i s . , @,,d, = I([ketonelc,,,,,,, + [M-COlc,,~,.)/[ketonelf,,,,,,J. (6) Maguire, J. A.; Boese, W. T.; Goldman, A. S . J . Am. Chem. Soc. 1989, 111, 7088.(7) Herkstroeter, W. G.; Lamola, A. A.; Hammond, G. S. J . Am. Chem. Soc. 1964, 86, 4537. (8) Turro, N. J. Modern Molecular Phofochemisfry;Benjamin Cummings: Menlo Park, CA, 1978; pp 362-368. (9) Reference 8, p 385. ( I O ) Cyclohexane photocarbonylation occurs to a slight extent in solutions of aromatic ketones in the absence of metal carbonyl under 800 Torr of C O (A = 366 nm, 0.10 M @-CF,C,,H,)C(O)CH,, GrC,= $,,dl = 0.0028). Yields = are significantly higher under higher pressures: e.g.. 400 psi of CO, $I,,~) = 0.077). Boese, W. T.; Goldman, A . S., to be published.

0 1992 American Chemical Society

J . Am. Chem. SOC.1992, 114, 351-353

The observations above suggest a mechanism of the general form of eqs 2-5. CyH

+ ArR'C=O A Cy' + ArR'C(0H)' Cy'

+ M(C0)

-

'M[C(O)Cy]

( M = RhL,Cl, Ru(CO),(dmpe), Ir(PMeJ,(CO)CI) 'M[C(O)Cy]

+ ArR'C(0H)'

HM[C(O)Cy]

+ CO

-

(2)

HM[C(O)Cy]

M(C0)

(3)

+ ArR'C=O (4)

+ HC(0)Cy

(5)

Equation 2 is well studied; known quantum yieldsI3of ca. 0.5 are consistent with the quantum yields of eq 1 calculated on the basis of the fraction of incident light absorbed by the ketone, dad, I 0.6. Equation 4 is well precedented by @-hydrogen atom transfers from organic radicals to metal radicals which exhibit near diffusion controlled kinetic^.'^ Reaction 3 may proceed via initial attack of Cy' on the metal center",'(' or, alternatively, by direct attack on coordinated CO. Note that the attack of Cy' on free CO may be considered an unlikely pathway (under 1 atm of CO1") on the basis of the slow rates known for radical addition to C0.I: A steady-state analysis, in which the rate of Cy' formation is assumed greater than or equal to the total rate of Cy-containing products,I8predicts that the ratio of formation of CyCHO/Cy: would be less than 0.4:1 as compared with an experimental value of 95:l (2.0 mM 2, 0.10 M acetophenone). The viability of eq 3 was confirmed by experiments in which cyclohexyl radicals were independently generated: irradiation (A > 420 nm) of Mn2(CO)lo(1.3 mM) in the presence of CyBr (1.6 M) and 2 (21 mM) resulted in the formation of Mn(CO),Br (1.5 mM) and Ir(PMe3)2[C(0)Cy]ClBr(8 mM); 1 reacted similarly. Presumably these reactions proceed via abstraction of Br' by 'Mn(C0)5,20followed by eq 3, and the resulting acylmetalloradical then abstracts Br' from CyBr (propagating a chain reaction). Aldehyde yields up to 110 mM have been realized from eq 1 (0.10 M acetophenone, 2.0 mM 2). In a separate experiment, 9.5 catalytic turnovers based on acetophenone (6.0 mM) were obtained (2.0 mM 2). Aldehyde decarbonylation (evidenced in experiments with added 0.1 M cyclooctanecarboxaldehyde) and ketone decomposition are yield-limiting factors. In summary, we report that photochemical carbonylation of cyclohexane is catalyzed by various d8 transition metal carbonyls and aromatic ketones. The mechanism involves cleavage of the ( 1 1) (a) Nomura, K.; Saito, Y . J . Chem. Soc., Chem. Commun. 1988, 161. (b) Sakakura, T.: Sodeyama, T.; Tokunaga, Y.: Tanaka, M. Chem. Lett. 1988, 263. (12) 2 is generated in situ from Ir(PMe,),(CO)CI under a C O atmosphere. The equilibrium constant for C O addition is 14.6 atm-' at 25 OC; thus 9.6% of the iridium is present as dicarbonyl 2 under 2 atm of CO. The rate dependence on C O pressure for iridium-catalyzed eq 1 is similar to that found for 1 (which does not add C O even under 1000 psi). This argues against the likelihood of Ir(PMeJ2(CO)CI being the active species since its concentration, unlike [ I ] , is CO-pressure dependent. ( I 3) (a) Beckatt, A,; Porter, G. Trans. Faraday SOC.1963, 59, 2038. (b) Reference 8, p 383. (14) (a) Halpern, J. Pure Appl. Chem. 1986, 58(4), 575. (b) Halpern, J. In Fundamental Research in Homogeneous Catalysis; Tsutsui, M., E,d.; Plenum Publishing: New York, 1979: Vol. 3, pp 25-40 and references therein. (15) Alkyl radical attack at 16-electron metal centers is precedented: Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of Organotransition Metal Chemistry; University Science Books: Mill Valley, CA, 1987; pp 314-315. (16) For radical attack at an 18-electron metal center, see: Brunet, J.; Sidot, C.; Caubere, P. J . Organomet. Chem. 1980, 204, 229. ( I 7) For example, for addition of methyl radical to CO, k = 3.0 X IO' M-' 5.' at 7 "C: Watkins, K. W.; Word, W. W. Int. J . Chem. Kinet. 1974, 6, 855. (18) The following rate constants were used: for addition of cyclohexyl radical to CO, k = 1.0 X 10' M - ' s ~ ' for ; dimerization of cyclohexyl radical, k = 1.5 X IO' M ' s-' (ref 19: see supplementary meterial). (19) Carlsson, D. J.; Ingold, K. U . J . A m . Chem. SOC.1968, 90, 7047. (20) Geoffroy, G. L.; Wrighton, M. S. Organometallic Photochemistry; Academic Press: New York, 1979; pp 136-140.

351

cyclohexane C-H bond by the photoexcited ketone and attack of the resulting cyclohexyl radical on the metal carbonyl.

Acknowledgment. This work was supported by the Division of Chemical Sciences, Office of Basic Energy Sciences, Office of Energy Research, U S . Department of Energy. A.S.G. thanks the Dreyfus Foundation for a Distinguished New Faculty Grant. We thank Zheng Wang for the synthesis of R ~ ( C O ) ~ ( d m pand e) Johnson-Matthey for generous loans of rhodium and iridium. Supplementary Material Available: Table of data of 28 carbonylation experiments, calculation of the upper limit of the ratio of CyCHO/Cy, formation based on the assumption of rate-limiting attack by Cy' on free CO, experimental data for the formation of Cy, and other side products in reactions with 0.0-4.0 mM 2, and synthesis and characterization of Ir(PMe,),[C(O)CylClBr ( 5 pages). Ordering information is given on any current masthead page.

Enthalpy Measurements in Organic Solvents by Photoacoustic Calorimetry: A Solution to the Reaction Volume Problem Rebecca R. Hungt and Joseph J. Grabowski*.'

Departments of Chemistry, Harvard University Cambridge, Massachusetts 021 38 University of Pittsburgh, Pittsburgh, Pennsylvania 15260 Received August 28, 1991 Photoacoustic calorimetry (PAC) has recently become an established tool for the determination of enthalpies of photoinitiated processes.' For the majority of work to date, it has been assumed that the photoacoustic signals detected in organic solvents arise exclusively from the thermal relaxation of the photoexcited chromophore. While this assumption leads to essentially no error for systems with no net photochemistry, differences between the partial molar volumes of the reactants and products should not be ignored for many studies in nonaqueous solution. In particular, photofragmentations and systems with large, photoinduced changes in polarity are subject to significant errors if the reaction volume is not explicitly considered. Here, we report a direct method, using a homologous series of alkane solvents, to resolve the thermal and the reaction volume contributions to the photoacoustic signal, and thus to improve the accuracy of the enthalpies determined by PAC. We examined the photodissociation of diphenylcyclopropenone (DPCP), which yields diphenylacetylene (DPA) and carbon monoxide (eq 1). Because of the unique properties of cyclo-

Ap3371 nm solvent

P h C S P h + CO

(1)

Ph

propenones, there have been several thermochemical studies on DPCP,*s3 including two widely different determinations of

'Harvard University.

*University of Pittsburgh. (1) Rothberg, L. J.; Simon, J . D.; Bernstein, M.; Peters, K. S. J . Am. Chem. Soc. 1983,105,3464-3468. Peters, K. S. Pure Appl. Chem. 1986.58, 1263-1266. Burkey, T. J.; Majewski, M.; Griller, D. J . Am. Chem. Soc. 1986, 108,2218-2221. Peters, K. S.; Snyder, G. J. Science 1988, 241, 1053-1057. Ni, T.; Caldwell, R. A.; Melton, L. A. J . Am. Chem. SOC.1989,111,457464. Kanabus-Kaminska, J. M.; Gilbert, B. C.; Griller, D. J . Am. Chem. Soc. 1989, 111, 3311-3314. Griller, D.; Wayner, D. D. M. PureAppl. Chem. 1989,61, 717-724. Morse, J. M., Jr.; Parker, G. H.; Burkey, T. J. Organometallics 1989,8, 2471-2474. Burkey, T . J. Polyhedron 1989, 8, 2681-2687. (2) Bostwick, D.; Henneike, F. H.; Hopkins, H . P., Jr. J . Am. Chem. SOC. 1975, 97, 1505-1509. Hopkins, H. P., Jr.; Bostwick, D.; Alexander, C. J . J . A m . Chem. SOC.1976, 98, 1355-1357. Greenberg, A,; Tomkins, R. P. T.; Dobrovolny, M.; Liebman, J . F. J . A m . Chem. Soc. 1983, 105, 6855-6858. Davis, H. E.; Allinger, N . L.; Rogers, D. W. J . Org. Chem. 1985, 50, 360 1-3604.

0002-7863/92/ 15 14-351%03.00/0 0 1992 American Chemical Society